Link to the single-page HTML spec instead of multipage - #63375

import re, subprocess, collections
RE = re.compile(r"(https?://)?html\.spec\.whatwg\.org/multipage(?:/multipage)?(?:/([A-Za-z0-9._-]*))?(#[^\s\"'<>`]*)?")
files = subprocess.check_output(["git", "grep", "-l", "html.spec.whatwg.org/multipage", "--", ":!tools/lint"], text=True).split("\n")
c = collections.Counter()
for f in filter(None, files):
    s = open(f, encoding="utf-8", errors="surrogateescape").read()
    for m in RE.finditer(s):
        if not m.group(3):
            c[(m.group(2), f)] += 1
for k, v in sorted(c.items()):
    print(v, k)

import json, re, subprocess, sys
ids = json.load(open(sys.argv[1]))
special = {"": "", "index": "", "popover": "#the-popover-attribute",
           "structured-data": "#safe-passing-of-structured-data"}
RE = re.compile(r"html\.spec\.whatwg\.org/multipage(?:/multipage)?(?:/([A-Za-z0-9-]*)(?:\.html?)?)?/?(?=(#?))")

def repl(m):
    page = m.group(1) or ""
    if m.group(2):
        return "html.spec.whatwg.org/"
    if page in special:
        return "html.spec.whatwg.org/" + special[page]
    if page in ids:
        return "html.spec.whatwg.org/#" + page
    raise Exception(f"unknown page {page!r}")

files = subprocess.check_output(["git", "grep", "-l", "html.spec.whatwg.org/multipage", "--", ":!tools/lint"], text=True).split("\n")
for f in filter(None, files):
    with open(f, "rb") as fh:
        s = fh.read().decode("utf-8", "surrogateescape")
    try:
        t = RE.sub(repl, s)
    except Exception as e:
        print(f, e)
        continue
    if t != s:
        with open(f, "wb") as fh:
            fh.write(t.encode("utf-8", "surrogateescape"))